As a strategic tool to reduce costs and increment the availability of the plant, Preventive Maintenance has become quite popular over the last years due to the significant increase of its application in industries. Despite the great quantity of models that deal with that problem, these are less than satisfactory in many situations when failure data are inexistent or not too reliable and when the observation of the conduct of more than one aspect is desirable to establish the periodicity of Preventive Maintenance. In such context, this article offers a model of decision by making use of the multicriteria method PROMETHE II and the Bayesian analysis, which allows to establish the periodicity of the Preventive Maintenance, observing the costs and reliability parameter and also permitting the lack of data not to be an impediment in the stipulation of those timings.
multicriteria decision; preventive maintenance; maintenance policies
